{
  "id": 13849,
  "title": "Is the prediction count actually 4700?",
  "url": "/competitions/facebook-recruiting-iv-human-or-bot/discussion/13849",
  "author_name": "",
  "post_date": "2015-04-29T11:38:00.240Z",
  "votes": -1,
  "comment_count": 3,
  "views": 1166,
  "content": "<p>I'm getting a result of human bidders as&nbsp;6512 out of&nbsp;7656335 bids. Total robot bids were found to be&nbsp;7649823 considering the single constraint mentioned to identify robots. &nbsp;The constraint is just bidder_id against the outcome(1.0) of trained data. The total amount of time taken to complete execute is exactly 4 minutes.</p>",
  "messages": [
    {
      "id": "75396",
      "postDate": "04/29/2015 11:38:00",
      "content": "<p>I'm getting a result of human bidders as&nbsp;6512 out of&nbsp;7656335 bids. Total robot bids were found to be&nbsp;7649823 considering the single constraint mentioned to identify robots. &nbsp;The constraint is just bidder_id against the outcome(1.0) of trained data. The total amount of time taken to complete execute is exactly 4 minutes.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "75425",
      "postDate": "04/29/2015 15:22:32",
      "content": "<p>I do not fully understand your question or the&nbsp;relation with its&nbsp;execution time, but, if you want to know the &quot;predictions count&quot;, if I understand correctly what you meant, just check the test.csv you need to have as much as you find in there.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "75487",
      "postDate": "04/29/2015 19:51:02",
      "content": "<p>The test count is 4700 bidders, while each bidder has 1 or more bids, to merge (1) test's bidders &quot;test.csv&quot; and (2) bids &quot;bids.csv&quot;, would result in much more than 4700 rows. (one to many)</p>\n<p>I guess.... the question wants to know whether prediction should be made for the 4700 rows or the merged test. </p>\n<p>CMIIW.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "75499",
      "postDate": "04/29/2015 21:06:02",
      "content": "<p>From sample submission file it's clear that the prediction is to be made for 4700 rows.</p>\n<p>Have a look at sample submission file it will be more clearer to you what to predict</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 75425,
      "author_name": "carloshuertas",
      "author_url": "",
      "post_date": "04/29/2015 15:22:32",
      "content": "<p>I do not fully understand your question or the&nbsp;relation with its&nbsp;execution time, but, if you want to know the &quot;predictions count&quot;, if I understand correctly what you meant, just check the test.csv you need to have as much as you find in there.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 75487,
      "author_name": "aldente",
      "author_url": "",
      "post_date": "04/29/2015 19:51:02",
      "content": "<p>The test count is 4700 bidders, while each bidder has 1 or more bids, to merge (1) test's bidders &quot;test.csv&quot; and (2) bids &quot;bids.csv&quot;, would result in much more than 4700 rows. (one to many)</p>\n<p>I guess.... the question wants to know whether prediction should be made for the 4700 rows or the merged test. </p>\n<p>CMIIW.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 75499,
      "author_name": "sahildahiya",
      "author_url": "",
      "post_date": "04/29/2015 21:06:02",
      "content": "<p>From sample submission file it's clear that the prediction is to be made for 4700 rows.</p>\n<p>Have a look at sample submission file it will be more clearer to you what to predict</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "75396": "",
    "75425": "",
    "75487": "",
    "75499": ""
  },
  "source": "meta"
}